[1-11C]Acetate PET


Acetate is quickly transported into cells by monocarboxylate transporters, and then converted into acetyl-CoA by acetyl-CoA synthetases in the cytosol and mitochondria. Acetyl-CoA can either enter into the TCA cycle in mitochondria (the main route in myocardium and brain), or it is incorporated into structural lipids (mainly in tumours) by fatty acid synthetase or acetyl-CoA carboxylase (Grassi et al., 2012). Acetyl-CoA can also enter into synthesis of amino acids (mainly glutamate and glutamine) and cholesterol, and used in gluconeogenesis. In the brain acetate is predominantly metabolized by astrocytes.

Acetate concentration in the blood plasma is usually <100 µM, higher after feeding, and can increase in liver diseases, acidosis, and after alcohol consumption. Acetate concentration depends on the sampling site, sample preparation, and storage (Tollinger et al., 1979; Scutches et al., 1979).


C-11 labeled acetic acid

[1-11C]acetate (Pike et al., 1982) is used to assess oxidative metabolism and perfusion.

Acetate molecule has two carbon atoms, either or both of which could be replaced with 11C. In vivo kinetics of [11C]acetate depend on which position the radionuclide is labelled. The label from the C-l position (carboxyl group) is released as [11C]CO2 in the second pass through the TCA cycle. The label from C-2 position (methyl group) can happen only after 2 passes, with ∼50% probability in each pass; thus much larger proportion of 11C in [2-11C]acetate would be incorporated in amino acid pool (van den Hoff et al., 1996; Klein et al., 2001).

After i.v. injection of [1-11C]acetate, the build-up phase of radioactivity in the tissue is related (but not linearly) to tissue perfusion. Washout of radiolabel represents the formation of [11C]CO2 in tissue, and is therefore related to oxygen consumption (Buck et al., 1991; Klein et al., 2001). Slower washout at later time points may represent the rate of fatty acid synthesis (Lewis et al., 2014).

The pancreas, bowel, kidneys, and spleen receive the highest radiation doses, but no urinary excretion can be detected (Seltzer et al., 2004). Dosimetry of [1-11C]acetate is favourable, because the main metabolite, [11C]CO2, is rapidly exhaled.

Input function

For quantitative analysis, arterial sampling should be performed, because concentrations of the parent tracer and metabolites differ between arterial and venous blood. Input function can often be derived from dynamic image with reasonable accuracy, especially in cardiac studies, but also if the arch of aorta, abdominal aorta, or iliac arteries are located in the image. Only BTAC can be derived from the image, and separate blood samples are required for conversion of BTAC to PTAC and metabolite correction; alternatively, population-based corrections can be applied.

Blood versus plasma

In a dog study, [1-14C]acetate concentration in red blood cells was negligible at least for 2 hours (Persson et al., 1991). Plasma-to-blood ratio in a [1-11C]acetate mice study (Authier et al., 2008) was about 1.45 at 5 min p.i., and the ratio then decreased to 1.35 at 15 min, 1.32 at 30 min, and 1.18 at 60 min, suggesting that unchanged [1-11C]acetate stays in plasma, but radioactive metabolites can penetrate the red blood cell membrane. Physiologically, though, acetate concentration in the human blood is higher than in plasma (Tollinger et al., 1979).

Radioactive metabolites

The main metabolite of [1-11C]acetate in blood is [11C]CO2 (Buck et al., 1991). Despite of the fact that [11C]CO2 is rapidly exhaled, it is present in the blood, and need to be accounted for in quantitative analysis. In humans, 10 min after administration ∼60% of the blood activity is due to [11C]CO2, and less than 20% due to [1-11C]acetate (Buck et al., 1991; Sun et al, 1998). The metabolite fractions plotted as a function of time clearly show a sigmoidal pattern (Buck et al., 1991, Fig 5; Sun et al., 1997, Fig 4; Sun et al, 1998, Table 3). Buck et al. (1991) did not fit the measured fractions nor directly used those in the data analysis, but instead the metabolite correction was included in the compartmental model for myocardium. The metabolite fractions were modelled using exponential function

, and when fitted together with the myocardial data, provided estimates a0=0.91±0.11 and μ=5.3±1.2 min. The same method was used in a rat study by Croteau et al (2010 and 2012), and a similar approach, but fitting several ROIs simultaneously, was used by Raylman et al (1994). The model-derived estimates of metabolite fractions were found to be fairly close to the measured fractions (Buck et al., 1991, Fig 5). This function with the reported mean parameter values has been since used for metabolite correction (van den Hoff et al., 1996 and 2001; Wyss et al., 2009). Schiepers et al. (2008) used a modified version of this function, where metabolite fraction starts to increase after a delay time:

, where the delay time τ=0.48 min and λ=0.104. In a dog study, marked blood concentration of [11C]CO2 was seen at ∼4 min (Brown et al., 1988, figure 1). Better approach would be to use the fractions reported by Sun et al (1998), because the values are based on direct blood measurements, independent on the tissue analysis model, and they assessed also other metabolites than [11C]CO2; these values have since been used for metabolite correction in a few studies (Timmer et al., 2011; Wong et al (2013); Hansson et al., 2017; Harms et al., 2018).

[C-11]Acetate parent fractions

Figure 1.[1-11C]acetate fractions in arterial blood (Sun et al., 1998), with fitted Hill function. Hill function parameters: 0.09957, 2.305, 26.58, 1.0, 0.0.

When [1-11C]acetate is used for estimating perfusion, single-tissue compartmental model can be fitted to only the first 3 min of the scan data, obviating the need for metabolite correction (Sciacca RR et al., 2001).


Bolus injection of [1-11C]acetate into coronary artery leads to high extraction and initially monoexponential clearance curve, which very well corresponds to myocardial oxygen consumption. Intravenous administration leads to dispersed input function, containing recirculating [1-11C]acetate and its labelled metabolites. This affects the myocardial TAC shapes, and leads to biased results unless kinetic model with measured input function is used (Buck et al., 1991). However, graphical methods for estimating the release rate of [11C]CO2 are attractive because of their simplicity; also, clearance method is independent on partial volume effect. Clearance of [11C]-acetate from the myocardium has been found to be bi-exponential (Brown et al., 1987, 1988 and 1989; Armbrecht et al., 1989). The k1 from bi-exponential and kmono from mono-exponential clearance estimation are correlated to myocardial oxygen consumption (Buxton et al, 1989; Walsh et al., 1989; Armbrecht et al, 1989 and 1990; Ng et al., 1994; Sun et al, 1998; Porenta et al., 1999; Ukkonen et al., 2001; Wong et al., 2013). Parametric kmono and perfusion images can be computed and presented as polar maps (Kotzerke et al., 1990; Miller et al., 1990; Hussain et al., 2009; Croteau et al., 2015).

Bi- and monoexponential functions that have been used to fit the decreasing myocardial TAC include

, where parameters k1, kc, and kmono correlate with the rate of oxidative metabolism and parameter k2 from compartmental model. Exponential fit is started at the time of onset of the most rapid decline in the TAC (Brown et al., 1988; Buck et al., 1991), not from the peak of the TAC. In practise, the monoexponential fitting of kmono has often been done by fitting line to the linear portion of semilogarithmic plot of the data. In a rat study, semilogarithmic plot was approximately linear 2-20 min p.i. in normoxic and hypoxic hearts and 10-35 min in ischemic hearts (Ng et al., 1994). In human studies, Timmer et al. (2011) started the monoexponential fit from the first frame for which four consecutive following frames showed decreasing activity concentrations in the whole myocardium TAC. Hansson et al. (2018) started fit at 6 min, which provided kmono with as good repeatability as k2 from single-tissue compartmental model, although kmono values were lower. Alternatively, clearance rate constant can be calculated via mean transit time; this method is less sensitive to the shape of the input function and avoids subjective selection of the linear portion of the data used for fitting (Choi et al., 1993).

Several compartmental models have been presented to estimate myocardial oxygen consumption, as reviewed by Klein et al. (2001). One-tissue compartment model analysis of [1-11C]acetate data allows also quantification of myocardial perfusion at rest as well as under stress conditions (van den Hoff et al., 2001; Sciacca RR et al., 2001; Sörensen et al., 2010). This model is a simplification of previous five-compartment model (van den Hoff et al., 1996), and performed best in comparison to three other models in assessment of MBF (Timmer et al., 2010). It was found to provide MBF values in fairly good agreement with actual perfusion values in both healthy individuals and patients with hypertrophic cardiomyopathy over physiological flow ranges under baseline conditions (Sciacca et al., 2001; Timmer et al., 2010), and a reliable index (k2) of oxygen consumption (Timmer et al., 2011; Wong et al., 2013). Arterial blood curve is extracted from left ventricular region, and corrected for metabolism using population-based function. Clustering can be applied in deriving the blood curve from the LV cavity (Harms et al., 2015). Basis function method can be used to calculate parametric images (Harms et al., 2018).

[1-11C]acetate PET has been used to assess myocardial efficiency by measuring both oxygen consumption and stroke volume from the single PET study (Sörensen et al., 2003 and 2010) with good repeatability (Hansson et al., 2018; Wu et al., 2018). Myocardial mass and volume can be calculated by segmenting parametric K1 and VB images (Harms et al., 2016) or from gated uptake images (Hansson et al., 2016).

Carimas™ includes the one-exponential fitting (kmono) for assessing myocardial oxygen consumption (Nesterov et al., 2015), and one-tissue compartmental model for the estimation of myocardial perfusion (van den Hoff et al., 2001; TPCMOD0039). Carimas™ user documentation contains further assistance on using the software.


Urinary excretion of acetate and its labelled metabolites is negligible, and therefore [1-11C]acetate is better for imaging tumours of the bladder and prostate cancer than for example [18F]FDG. In a prospective study, [1-11C]acetate PET/MRI was shown to be feasible for staging of bladder cancer (Salminen et al., 2018).

[1-11C]acetate has high uptake in renal parenchyma, and there are conflicting results on whether it is not useful for detecting renal cell carcinoma (Kotzerke et al., 2007; Oyama et al., 2009).

Prostate cancer

Schiepers et al. (2008) observed that MTGA for irreversible uptake (Patlak plot) and 2-tissue compartment model, with k4 set to zero, can be used to study the metabolic activity of prostate tumours. Blood metabolites were corrected using previously estimated metabolite function. SUV will be sufficient in clinical practice and IMRT treatment planning (Schiepers et al., 2008; Seppälä et al., 2009).


Total cerebral oxygen consumption can be measured using inhaled [15O]O2, but astrocytic oxidative metabolism can be measured using [1-11C]acetate (Wyss et al., 2009; Iversen et al., 2014; Arnold et al., 2015). Wyss et al. (2009) used traditional one-tissue compartment model fitting to estimate K1 and k2, with blood volume fraction fixed to 0.05. Metabolite corrected arterial plasma TAC was used as model input: measured blood curves were converted to plasma curves using quadratic polynomial function, which was fitted to plasma/blood-ratios obtained from rat studies (Wyss et al., 2009); metabolite correction was based on previously published metabolite fractions in humans (Buck et al., 1991). K1 was weakly correlated with perfusion (because of low extraction), but k2 seemed to be more correlated with oxygen metabolism than perfusion.

Iversen et al. (2014) developed a three-tissue compartmental model, in which parameter k3 represents the oxidation rate of [11C]acetate. Non-linear mixed effects model was used to fit all study subjects (including three separate groups) simultaneously to avoid overfitting.


Dynamic PET studies with [1-11C]acetate can be analyzed using arterial input function and one-tissue compartmental model (Shreve et al., 1995). PET images are of good quality even in case of severely reduced renal function. Acetate is resorbed from glomerular filtrate by active transport in the proximal convoluted tubules (Schafer & Williams, 1985), and therefore the tracer has no observable urinary excretion at least during the first 30 min after tracer administration. Extraction of [1-11C]acetate is ∼20-25%. Both K1 and k2 are reduced in renal disease and renal artery stenosis (Shreve et al., 1995). Juillard et al (2007) showed in a pig study that kmono can be calculated from a mono-exponential fit and that it correlates well with renal oxidative metabolism. Acetate can also be labelled with 13C, and used to study the oxidative metabolism of kidneys with hyperpolarized MRI (Mikkelsen et al., 2017); diuretic furosemide induced change in oxygen consumption could be demonstrated in rats using both [1-11C]acetate-PET and [1-13C]acetate-MRI.


In liver studies the portal and arterial contribution to the input function can be taken into account in the analysis method (Chen and Feng, 2004a and 2006; Chen et al. 2004b). For diagnostic purposes, however, SUV or liver-to-blood ratio (LBR) can be sufficient; for instance, both SUVmax and LBR have been shown to detect fatty infiltration in liver steatosis with high accuracy (Nejabat et al., 2018). Sensitivity for detecting hepatic tumours is poor (Roivainen et al., 2013).

Normal SUV in liver, pancreas, spleen, and adrenal glands have been reported by Malkowski et al (2017).

Skeletal muscle

[1-11C]acetate can be used to calculate indices of muscle blood flow and oxidative metabolism, applying two-tissue compartmental model with metabolite corrected plasma input (Croteau et al., 2010; Labbe et al., 2011). The rate constant K1 provides a reliable index of tissue perfusion, because the first-pass extraction fraction of [1-11C]acetate is close to 1 in the resting skeletal muscle. The rapid component of tissue clearance, rate constant k2, is assumed to represent muscle oxidative metabolism (Labbe et al., 2011). Slow [11C]CO2 release from resting muscle may limit the reliability of estimates of oxidative metabolism.

Buchegger et al (2011) have analyzed [1-11C]acetate uptake in resting and exercising muscle semi-quantitatively using SUV.

Adipose tissue

Mono-exponential clearance (kmono) has been used as an estimate of oxidative metabolism in human brown adipose tissue (BAT) studies (Ouellet et al., 2012; Blondin et al., 2014, 2015a, 2015b, 2017a, and 2017b). In all of these studies the monoexponential fit was started from the TAC peak. In white adipose tissue (WAT) [1-11C]acetate uptake is very low (Ouellet et al., 2012) and clearance cannot be measured using exponential fit. Also in BAT in normal conditions the uptake is low and clearance very slow, and estimation of kmono may be prone to errors.

Irreversible two-tissue compartmental model, developed for myocardium (van den Hoff et al., 1996), has been used to estimate both perfusion (K1) and oxidative metabolism (k2) in the BAT of rats (Labbé et al., 2015) and BAT and WAT of mice (Labbé et al., 2018).


Early pulmonary retention of [1-11C]acetate has been measured as SUV; lung water content was quantitated from [15O]H2 PET at equilibrium (Sörensen et al., 2006). Monoexponential fit, started 5 min p.i., has been used to calculate kmono (Sörensen et al., 2006).

See also:


Armbrecht JJ, Buxton DB, Brunken RC, Phelps ME, Schelbert HR. Regional myocardial oxygen consumption determined noninvasively in humans with [1-11C]acetate and dynamic positron tomography. Circulation 1989; 80: 863-872. doi: 10.1161/01.CIR.80.4.863.

Armbrecht JJ, Buxton DB, Schelbert HR. Validation of [1-11C]acetate as a tracer for noninvasive assessment of oxidative metabolism with positron emission tomography in normal, ischemic, postischemic, and hyperemic canine myocardium. Circulation 1990; 81: 1594-1605. doi: 10.1161/01.CIR.81.5.1594.

Arnold A, Calvetti D, Gjedde A, Iversen P, Somersalo E. Astrocytic tracer dynamics estimated from [1-11C]-acetate PET measurements. Math Med Biol. 2015; 32(4): 367-382. doi: 10.1093/imammb/dqu021.

Authier S, Tremblay S, Dumulon V, Dubuc C, Ouellet R, Lecomte R, Cunnane SC, Bénard F. [11C] Acetoacetate utilization by breast and prostate tumors: a PET and biodistribution study in mice. Mol Imaging Biol. 2008; 10(4): 217-223.

Brown MA, Myears DW, Bergmann SR. Validity of estimates of myocardial oxidative metabolism with carbon-11 acetate and positron emission tomography despite altered patterns of substrate utilization. J Nucl Med. 1989; 30: 187-193.

Buck A, Wolpers HG, Hutchins GD, Savas V, Mangner TJ, Nguyen N, Schwaiger M. Effect of carbon-11-acetate recirculation on estimates of myocardial oxygen consumption by PET. J Nucl Med. 1991; 32: 1950-1957.

Buchegger F, Ratib O, Willi J-P, Steiner C, Seimbille Y, Zaidi H, Graf V, Peter R, Jung M. [11C]acetate PET/CT visualizes skeletal muscle exercise participation, impaired function, and recovery after hip arthroplasty; first results. Mol Imaging Biol. 2011; 13: 793-799. doi: 10.1007/s11307-010-0415-9.

Buxton DB, Nienaber CA, Luxen A, Ratib O, Hansen H, Phelps ME, Schelbert HR. Noninvasive quantitation of regional myocardial oxygen consumption in vivo with [1-11C]acetate and dynamic positron emission tomography. Circulation 1989; 79: 134-142. doi: 10.1161/01.CIR.79.1.134.

Chen S, Feng D. Noninvasive quantification of the differential portal and arterial contribution to the liver blood supply from PET measurements using the 11C-acetate kinetic model. IEEE Trans Biomed Eng. 2004a; 51(9): 1579-1585. doi: 10.1109/TBME.2004.828032.

Chen S, Ho C, Feng D, Chi Z. Tracer kinetic modeling of 11C-acetate applied in the liver with positron emission tomography. IEEE Trans Med Imaging 2004b; 23(4): 426-432. doi: 10.1109/TMI.2004.824229.

Chen S, Feng D. Novel parameter estimation methods for 11C-acetate dual-input liver model with dynamic PET. IEEE Trans Biomed Eng. 2006; 53(5): 967-973. doi: 10.1109/TBME.2006.872817.

Croteau E, Lavallée É, Labbe S, Hubert L, Pifferi F, Rousseau JA, Cunnane SC, Carpentier AC, Lecomte R, Bénard F. Image-derived input function in dynamic human PET/CT: methodology and validation with 11C-acetate and 18F-fluorothioheptadecanoic acid in muscle and 18F-fluorodeoxyglucose in brain. Eur J Nucl Med Mol Imaging 2010; 37: 1539-1550. doi: 10.1007/s00259-010-1443-z.

Grassi I, Nanni C, Allegri V, Morigi JJ, Montini GC, Castellucci P, Fanti S. The clinical use of PET with 11C-acetate. Am J Nucl Med Mol Imaging 2012; 2(1): 33-47.

Gropler RJ, Siegel BA, Geltman EM. Myocardial uptake of carbon-11-acetate as an indirect estimate of regional myocardial blood flow. J Nucl Med. 1991; 32(2): 245-251.

Han C, Oikonen V. Model equations for myocardial perfusion studies with [1-11C]acetate PET. TPCMOD0039.

van den Hoff J, Burchert W, Wolpers HG, Meyer GJ, Hundeshagen H. A kinetic model for cardiac PET with [1-carbon-11]-acetate. J Nucl Med. 1996; 37: 521-529.

van den Hoff J, Burchert W, Börner A-R, Fricke H, Kühnel G, Meyer GJ, Otto D, Weckesser E, Wolpers H-G, Knapp WH. [1-11C]-acetate as a quantitative perfusion tracer in myocardial PET. J Nucl Med 2001; 42: 1174-1182.

Hussain R, Kudo T, Tsujikawa T, Kobayashi M, Fujibayashi Y, Okazawa H. Validation of the calculation of the clearance rate constant (kmono) of [11C]acetate using parametric kmono image for myocardial oxidative metabolism. Nucl Med Biol. 2009; 36(8): 877-882. doi: 10.1016/j.nucmedbio.2009.07.002.

Juillard L, Lemoine S, Janier MF, Barthez PY, Bonnefoi F, Laville M. Validation of renal oxidative metabolism measurement by positron-emission tomography. Hypertension 2007; 50:242-247. doi: 10.1161/HYPERTENSIONAHA.107.089607.

Klein LJ, Visser FC, Knaapen P, Peters JH, Teule GJJ, Visser CA, Lammertsma AA. Carbon-11 acetate as a tracer of myocardial oxygen consumption. Eur J Nucl Med. 2001; 28: 651-668. doi: 10.1007/s002590000472.

Kotzerke J, Hicks RJ, Wolfe E, Herman WH, Molina E, Kuhl DE, Schwaiger M. Three-dimensional assessment of myocardial oxidative metabolism: a new approach for regional determination of PET-derived carbon-11-acetate kinetics. J Nucl Med. 1990; 31(11): 1876-1883.

Labbé SM, Croteau E, Grenier-Larouche T, Frisch F, Ouellet R, Langlois R, Guérin B, Turcotte EE, Carpentier AC. Normal postprandial nonesterified fatty acid uptake in muscles despite increased circulating fatty acids in type 2 diabetes. Diabetes 2011; 60: 408-415. doi: 10.2337/db10-0997.

Lewis DY, Boren J, Shaw GL, Bielik R, Ramos-Montoya A, Larkin TJ, Martins CP, Neal DE, Soloviev D, Brindle KM. Late imaging with [1-11C]acetate improves detection of tumor fatty acid synthesis with PET. J Nucl Med. 2014; 55: 1144-1149.

Nesterov SV, Turta O, Han C, Mäki M, Lisinen I, Tuunanen H, Knuuti J. C-11 acetate has excellent reproducibility for quantification of myocardial oxidative metabolism. Eur Heart J Cardiovasc Imaging 2015; 16(5): 500-506. doi: 10.1093/ehjci/jeu289.

Ouellet V, Labbé SM, Blondin DP, Phoenix S, Guérin B, Haman F, Turcotte EE, Richard D, Carpentier AC. Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. J Clin Invest. 2012; 122(2): 545-552. doi: 10.1172/JCI60433.

Schiepers C, Hoh CK, Nuyts J, Seltzer M, Wu C, Huang S-C, Dahlbom M. 1-11C-acetate kinetics of prostate cancer. J Nucl Med. 2008; 49(2): 206-215. doi: 10.2967/jnumed.107.044453.

Shreve P, Chiao P-C, Humes HD, Schwaiger M, Gross MD. Carbon-11-acetate PET imaging in renal disease. J Nucl Med. 1995; 36(9): 1595-1601.

Sun KT, Chen K, Huang S-C, Buxton DB, Hansen HW, Kim AS, Siegel S, Choi Y, Müller P, Phelps ME, Schelbert HR. Compartment model for measuring myocardial oxygen consumption using [1-11C]acetate. J Nucl Med. 1997; 38: 459-466.

Sun KT, Yeatman A, Buxton DB, Chen K, Johnson JA, Huang S-C, Kofoed KF, Weismueller S, Czernin J, Phelps M, Schelbert HR. Simultaneous measurement of myocardial oxygen consumption and blood flow using [1-carbon-11]acetate. J Nucl Med. 1998; 39: 272-280.

Timmer SA, Lubberink M, Germans T, Götte MJ, ten Berg JM, ten Cate FJ, van Rossum AC, Lammertsma AA, Knaapen P. Potential of [11C]acetate for measuring myocardial blood flow: Studies in normal subjects and patients with hypertrophic cardiomyopathy. J Nucl Cardiol. 2010; 17(2): 264-275. doi: 10.1007/s12350-009-9181-y.

Timmer SA, Lubberink M, van Rossum AC, Lammertsma AA, Knaapen P. Reappraisal of a single-tissue compartment model for estimation of myocardial oxygen consumption by [11C]acetate PET: an alternative to conventional monoexponential curve fitting. Nucl Med Commun. 2011; 32(1): 59-62. doi: 10.1097/MNM.0b013e32834113b0.

Ukkonen H, Knuuti J, Katoh C, Iida H, Sipilä H, Lehikoinen P, Någren K, Voipio-Pulkki L-M. Use of [11C]acetate and [15O]O2 PET for the assessment of myocardial oxygen utilization in patients with chronic myocardial infarction. Eur J Nucl Med. 2001; 28(3): 334-339. doi: 10.1007/s002590000444.

Wong YY, Raijmakers P, van Campen J, van der Laarse WJ, Knaapen P, Lubberink M, Ruiter G, Vonk Noordegraaf A, Lammertsma AA. 11C-acetate clearance as an index of oxygen consumption of the right myocardium in idiopathic pulmonary arterial hypertension: a validation study using 15O-labeled tracers and PET. J Nucl Med. 2013; 54(8): 1258-1262. doi: 10.2967/jnumed.112.115915.

Wyss MT, Weber B, Tryer V, Heer S, Pellerin L, Magistretti PJ, Buck A. Stimulation-induced increases of astrocytic oxidative metabolism in rats and humans investigated with 1-11C-acetate. J Cereb Blood Flow Metab. 2009; 29: 44-56. doi: 10.1038/jcbfm.2008.86.

Tags: , , , , , ,

Created at: 2008-11-27
Updated at: 2018-12-08
Written by: Vesa Oikonen, Chunlei Han